This is an extended version of the article I wrote for the latest edition of Custom PC Magazine in the U.K. Once again, if you're looking for the sexiest PC Hardware Magazine in Europe (and one of the best in the world) you should get his magazine! It's full of the best hardware photography ever, and the quality of the paper is to die for.Toms Hardware just wrote an interesting article called The SSD Power Consumption Hoax.
While they have some interesting points, I would suggest that there is much more involved with evaluating the potential increase in battery life with Solid State Technology.
A Solid State Drive can affect battery life in two situations: 1) Poor Performance, and 2) High Idle Power.
You cannot estimate the power savings of a solid state drive simply by looking at the hard drive subsystem and then extrapolating based on datasheet power specifications. That methodology does not comprehend the work being performed, and performance impacts energy efficiency. If a solid state drive can complete a workload 2x faster than a hard drive, then the entire platform can enter a power efficient state sooner.
We suspect that one reason that Tom's measured worse solid state battery life than a 7200 RPM HDD is likely that the particular solid state drive performed worse than the 7200 RPM HDD.
If the solid state drive completed the workload later, then the platform consumed more power. It also appears Tom’s chose solid state drives that use FPGAs, and these devices probably have very high idle power.
So bottom line is that power efficiency must comprehend the work being performed (during the power measurement.) A better metric is power per IOs per second. I know of at least one solid state drive that consumes only 60mWatts during idle, and it consumed less than 100mW on average in Mobilemark.
...and uhhh, a mobile hard disk drive consumes between 1 and 1.5 watts during a Mobilemark run.
So, does this mean that Tom's Hardware was right? Perhaps with their specific benchmarks - but even unintentionally, it's a pretty glorified "lets be controversial" slightly narrow view of the world. I'm not going to tear it apart - but it seems the author assumes that all solid state drives are created equal. There is a profound difference in performance and power depending on the product architecture and design. Tom's itself reported up to a 10x span in solid state drive raw performance depending on vendor, so it's interesting that this author assumed they are generic.
To put it in perspective, even if the hard drive were removed entirely, it only represents ~5% of the total battery consumption. So even if we halve the consumption with solid state we'd typically only gain 10 minutes battery life (out of 3 hours). Therefore a solid state can have a slight improvement on battery life. So what's the point of even writing about it? Ugh, wasting my time.
Trust me, this is NOT the end of the story. Stay tuned for later this year when the solid state $@#* hits the fan and spinning disk heads start to roll. There are some new players in town and they are bringing the big guns.
